About

My Ph.D. thesis focuses on Reconstructing Hand Shape, Pose and Appearance for Tracking from Monocular Video, advised by Parag Chaudhuri. Specifically, my work encompasses personalized shape modeling, articulated pose estimation, 3D model-based tracking using non-linear optimization, and illumination-aware appearance reconstruction using Monte Carlo path tracing in a differentiable deferred rendering framework. My research interests include digitizing real-world environments capturing detailed geometry and appearance of static/dynamic scenes, and real-time tracking of rigid/articulated/non-rigid objects. During my PhD, I was an intern with the Graphics group at Adobe Research. Previously, I completed my B.Tech. in Computer Science and Engineering from Indian Institute of Technology Jodhpur where I developed MIST (Medical Image Segmentation Tool for reconstructing 3D bone from MRI data) in collaboration with All India Institute of Medical Sciences Jodhpur as part of my B.Tech. project.

Publications

Intrinsic Hand Avatar: Illumination-aware Hand Appearance and Shape Reconstruction from Monocular RGB Video
Pratik Kalshetti, Parag Chaudhuri
WACV 2024
Paper | Code | Video

Reconstructing Hand Shape and Appearance for Accurate Tracking from Monocular Video
Pratik Kalshetti
SIGGRAPH Asia Doctoral Consortium 2023
Paper | Poster | Video

Local Scale Adaptation to Hand Shape Model for Accurate and Robust Hand Tracking
Pratik Kalshetti, Parag Chaudhuri
Computer Graphics Forum (ACM SIGGRAPH/Eurographics Symposium on Computer Animation) 2022
Paper | Code | Video

Local scale adaptation for augmenting hand shape models
Pratik Kalshetti, Parag Chaudhuri
SIGGRAPH Posters 2022
Paper
2nd place in ACM Student Research Competition
Featured in ACM SIGGRAPH Blog

Unsupervised incremental learning for hand shape and pose estimation
Pratik Kalshetti, Parag Chaudhuri
SIGGRAPH Posters 2019
Paper
3rd place in ACM Student Research Competition

Antara: An Interactive 3D Volume Rendering and Visualization Framework
Pratik Kalshetti*,Parag Rahangdale, Dinesh Jangra, Manas Bundele, Chiranjoy Chattopadhyay
arXiv, 2018
Paper

An Interactive Medical Image Segmentation Framework using Iterative Refinement
Pratik Kalshetti*, Manas Bundele, Parag Rahangdale, Dinesh Jangra, Chiranjoy Chattopadhyay, Gaurav Harit, Abhay Elhence
Computers in Biology and Medicine 2017
Paper | Project
Accorded Honors status
Shortlisted for best BTech project

Other Projects

Fast Point Cloud Sampling using Graph Signal Processing
This project uses graph signal processing to sample a point cloud such that application-dependent features are preserved. Initially a graph is constructed from the point cloud. A sampling distribution is then computed using this graph and the desired application. More precisely, the application decides the type of graph filter to be used for computing the distribution. Finally this distribution is used to sample the point cloud.
Code

Fit Mesh to Pointcloud
The algorithm is expressed as energy minimization. The energy is written as a sum of squares that is then optimized using Levenberg-Marquardt. For this project, the initialization is provided manually, however this can be provided by a discriminative model. The key novelty is to jointly optimize over both model parameters and correspondences between observed data points and the model surface.
Code

Object Detection
Single Shot MultiBox architecture was used to solve this classification + regression problem and the variable dimensional output was handled using anchor boxes, resulting in real-time accurate detection. First publicly available implementation of YOLOv2 in Tensorflow with Eager Execution API from scratch.
Code

Volunteer Activities

  • Student volunteer at SIGGRAPH 2021, virtual
  • Student volunteer at SIGGRAPH 2018, Vancouver
  • Head, Public Relations and Hospitality, IGNUS 2015, IIT Jodhpur
  • Coordinator, Electronics Club, IIT Jodhpur (2013)